"Towards Electronic Journals" (Tenopir & King 2000a,b) analyzes the scholarly journal publishing industry and the influences upon it that affect subscription costs. This book documents the impact journals have on scientists and libraries, especially in light of the Internet. Research studies corroborate most of the book's assertions, most notably the ones involving commercial publisher profit margins.
1. It is rare when a book that deals with a timely aspect of librarianship methodically addresses the issue it sets out to investigate. More often than not, theories are simply tossed about with little in the way of supporting evidence. More frustrating are those occasions when no specific thesis is stated, so that the monograph fails to guide readers down a continual road of understanding. Like trial lawyers, Tenopir & King (2000a,b) open with a clearly stated purpose, and proceed to argue their case based on a wide array of evidence. "Towards Electronic Journals" is an empirical analysis of journal publishing and its impact on scientists and libraries.
2. The scholarly journal system in the United States is a $45 billion industry. The authors open with this fact in an attempt to convey the magnitude of this business, especially in regard to publishers. Many librarians believe a shift from paper to electronic distribution of scholarly journals should decrease journal costs. As noted by the title, the thrust of this work revolves around this notion, and the results are not what one would expect.
3. Much of the book's analyses separate academic scientists from industrial scientists, since a variance exists between the two groups in terms of journal use. For instance, the authors show that scientists in a university environment spend 180 hours per year reading scholarly journals, while non-academic scientists spend only 100 hours. Moreover, 36% of university scientists' readings are from personal subscriptions, compared to 24% for scientists in other organizations. These percentages are considerably lower than those in 1977, which were 60% and 72% respectively. When added to an overall average of eight inter-library loan requests per scientist per year, it's not difficult to see the strain placed on libraries in terms of scholarly journal availability. The decline in revenue because of decreased personal subscriptions directly impacts journal costs. Tenopir and King clearly show this trend through the use of statistics gathered over the past 20 years.
4. The amount of scientific knowledge doubles every 15 to 17 years. As the authors note, recent college graduates have only been made aware of a fraction of the information that will be disseminated during their careers. Journals are the chief mechanism for keeping abreast of advances in one's field. As a result, browsing scholarly journals is the chief source of discovery for scientists, despite a steady increase in pertinent literature identified through online abstracting and indexing (A & I) agencies. In fact, less than 1% of journal articles were identified through automated searches in 1977. In recent years, this number has grown to over 12%. It is surprising, however, that the plethora of available A & I services don't constitute a larger percentage of articles identified as useful. Scientists seem more interested in particular periodicals compared to a subject-specific approach to information retrieval. Such a dynamic has implications for vendors of aggregated databases since these appear optimized for providing a search mechanism rather than a browsable interface to their journal collections.
5. The majority of scientific literature read has been published within the past year. Only one-third of articles read are more than one year old, and a mere 5% of these are more than 15 years old. The authors describe a situation that holds huge implications in the electronic arena. Many electronic journal publishers provide holdings going back five or more years. If only 5% of articles read are more than 15 years old, by 2010, 95% of all articles read will be available online. Although the authors feel that print and electronic journals will co-exist into the near future, it is easy to envision a scenario ten years from now where online journals are the only means of access to periodical articles. Scholarly journals in print would only be held by the largest research libraries.
6. A strength of this work, and one that is much needed, is the space allocated to describing the scholarly journal publishing model. It's certainly important for librarians to understand the nature of this business, and Tenopir and King provide us with a detailed overview. When adjusted for inflation, the price of scholarly journals has increased 2.6 times from 1975-1995. This figure at first glance seems high. Yet the authors clearly describe the costs inherent in the industry and factors that affect price. For instance, it costs $1,545 to publish an article in a scholarly journal and $325 per page published. In 1995, there were 6,771 scholarly scientific journals published in the United States. Moreover, cancelled subscriptions have a direct impact on subscription costs. In 1975, scientists subscribed to an average of 5.8 journals per year. In 1995, however, scientists averaged only 2.7 subscriptions. Although there are more scientists today than in 1975, there are also more journals. Libraries are being relied on to pick up the slack.
7. The authors lay a solid foundation which leads to the text's most significant area: the effect electronic journals will have on an industry considered by many to be dominated by vultures. In the print publishing model, the greater number of journal issues published, the greater the savings for both publishers and subscribers. An entire chapter with various mathematical equations breaks down the cost of scientific scholarly journal publishing, which includes article publishing costs, non-article publishing costs, reproduction costs, distribution costs, and publishing support costs. An interesting figure among these models shows the break-even point for publishers; that is, the minimum price necessary to recover all costs associated with publishing a scholarly journal based on number of subscribers:
NUMBER OF COST PER SUBSCRIBERS SUBSCRIPTION
500 $775 1000 $404 2500 $181 5000 $107 10000 $70
8. Commercial publishers are responsible for the top end of these averages. They have the highest cost per subscriber ($441) and average journal price ($487). As a result, contempt caused by skyrocketing journal prices is targeted primarily at commercial publishers. The hope is that electronic publishing of scientific scholarly journals will lower escalating journal prices. As the authors note, however, savings per electronic subscription go down with increased circulation. An electronic journal with a circulation of 500 saves subscribers $25 when compared to a print subscription. For an e-journal with a circulation of 50,000, the savings is only $19. Libraries do slightly better, with an average savings of $70 per subscription for electronic-only journals. Unlike timeliness and the lack of page restrictions, lower price cannot be considered a great advantage of electronic journals.
9. Be that as it may, e-journals are proliferating. In 1991, there were but seven peer-reviewed electronic journals. In 1997, the number climbed to 1,049. In 1999, there were approximately 4,000 scholarly electronic journals. Accordingly, more scientists are using electronic journals. In 1998, 48% of faculty at research universities accessed an e-journal. A year later, this figure rose to 61%.
10. Library door counts are dropping, but scientists are using library-provided materials more than ever. In 1984, the average academic scientist used the library 77 times. From 1990-1993, academic scientists averaged 119 uses of the library. Increasingly, these uses include electronic services, chiefly electronic journals. All library clientele, not just scientists, expect online access to journals. Such pressures force libraries to purchase expensive e-journal collections while maintaining print subscriptions, since there are few, if any, guarantees concerning electronic journal archiving.
11. As the authors state, "costs of exclusively electronic publishing are less than paper, although not appreciably so." As they note, many publishing activities are common to both print and electronic publishing. In addition, electronic-only costs such as storage and software offset savings in the print-only areas of reproduction and distribution. The actual savings seem to vary according to the source. The American Mathematics Society concludes that electronic publishing costs 90% of paper publishing. The Canadian Journal of Communications has found a savings of 25%, Still others claim the savings to be as high as 75% over paper. If savings to publishers are this great, subscribers, and especially libraries, have yet to reap the financial benefit.
12. E-journal pricing arrangements vary from publisher to publisher. As a whole, the system is complex and tedious for librarians to monitor. Two well-established pricing structures are library-negotiated license agreements and consortia discount purchasing. Since many libraries are reluctant to replace print subscriptions with electronic, these costs are add-ons to budgets which fail to keep up with the inflation rate of journals. The authors contend that libraries maintain both print and electronic subscriptions because they fear losing some services. However, this fear, and others such as archiving and restrictive licensing agreements, are not the deterrents they once were. Scientists' pursuit of journals in electronic format requires libraries to be more forward looking and less conservative in making collection development decisions.
13. Tenopir and King warn of the financial risks that parallel publishing (publishing simultaneously in print and electronic formats) can have on the journal system. Publisher pricing strategies may force libraries into purchasing only one medium or the other, which can result in extremely high prices. The authors believe that an appropriate subscription charge is $40 more for print subscriptions than for electronic. This assertion means that scholarly journal prices would be based on the cost to produce the electronic version, not the print. This concept is revolutionary, and no doubt would be considered heresy at many publishing houses. Yet initiatives such as SPARC may force commercial publishers to reconsider their profit margins.
14. It seems somewhat unprofessional to write an unblemished review. Readers often feel that the reviewer simply glanced over the text in an uncritical fashion. If I'm guilty of anything, it's of being overly eager to read a work that succeeds in elucidating an incredibly complex issue. The years of research and writing that has resulted in this book are evident throughout. Carol Tenopir and Donald King, authorities in this field, raise all the pertinent questions and provide evidence as answers. "Towards Electronic Journals" provides readers with a comprehensive understanding of scholarly journal publishing. My highest recommendation is not praise enough.
Tenopir, Carol, and Donald W. King (2000a) Towards Electronic Journals: Realities for Scientists, Librarians, and Publishers. Washington, D.C.: Special Libraries Association. http://www.sla.org
Tenopir, Carol, and Donald W. King (2000b) Precis of: "Towards Electronic Journals" PSYCOLOQUY 11(084) ftp://ftp.princeton.edu/pub/harnad/Psycoloquy/2000.volume.11/ psyc.00.11.084.electronic-journals.1.tenopir http://www.cogsci.soton.ac.uk/cgi/psyc/newpsy?11.084